This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Barone, F. C. Articles by Feuerstein, G. Z. Search for Related Content PubMed PubMed Citation Articles by Barone, F. C. Articles by Feuerstein, G. Z. Pubmed/NCBI databases Substance via MeSH

(Stroke. 1997;28:1233-1244.)
© 1997 American Heart Association, Inc.

Articles

Tumor Necrosis Factor-

A Mediator of Focal Ischemic Brain Injury

F. C. Barone, PhD; B. Arvin, PhD; R. F. White, BS; A. Miller, MD; C. L. Webb, MS; R. N. Willette, PhD; P. G. Lysko, PhD; G. Z. Feuerstein, MD

From the Department of Cardiovascular Pharmacology, SmithKline Beecham Pharmaceuticals, King of Prussia, Pa, and Department of Neurology, Carmel Medical Center, Haifa, Israel (A.M.).

Correspondence to Frank C. Barone, PhD, Department of Cardiovascular Pharmacology UW2521, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd, PO Box 1539, King of Prussia, PA 19406. E-mail Frank_C_Barone{at}SBPHRD.DOC

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Background and Purpose Tumor necrosis factor- (TNF-) is a pleiotropic cytokine that rapidly upregulates in the brain after injury. The present study was designed to explore the pathophysiological significance of brain TNF- in the ischemic brain by systematically evaluating the effects of lateral cerebroventricular administration of exogenous TNF- and agents that block the effects of TNF- on focal stroke and by examining the potential direct toxic effects of TNF- on cultured neurons to better understand how TNF- might mediate stroke injury.

Methods TNF- (2.5 or 25 pmol) was administered intracerebroventricularly to spontaneously hypertensive rats 24 hours before permanent or transient (80 minutes and 160 minutes) middle cerebral artery occlusion (MCAO). Animals were examined 24 hours later for neurological deficits and ischemic hemisphere necrosis and swelling. In some of these studies, neutralizing anti–TNF- monoclonal antibody (mAb) (60 pmol) was injected intracerebroventricularly 30 minutes before exogenous TNF- (25 pmol). In addition, the effects of blocking endogenous TNF- on permanent focal ischemic injury were determined with the use of either mAb (60 pmol) or soluble TNF receptor I (sTNF-RI) (0.3 or 0.7 nmol) administered intracerebroventricularly 30 minutes before and 3 and 6 hours after MCAO. Finally, the direct neurotoxic effects of TNF- were studied in cultured rat cerebellar granule cells exposed to TNF- (10 to 2000 U/mL for 6 to 24 hours), and neurotransmitter release, glutamate toxicity, and oxygen radical toxicity were studied.

Results TNF- increased the percent hemispheric infarct induced by permanent MCAO in a dose-related manner from 13.1±1.3% (vehicle) to 18.9±1.7% at 2.5 pmol (P<.05) and 27.1±1.3% at 25 pmol (P<.0001). The high dose of TNF- increased ischemia-induced forelimb deficits from 1.6±0.2 to 2.3±0.2 (P<.01). TNF- (2.5 pmol) also increased the infarction induced by 80 or 160 minutes of transient MCAO from 1.9±0.9% to 4.3±0.4% (P<.01) and from 14.2±1.3% to 21.6±2.2% (P<.05), respectively. The exacerbation of infarct size, swelling, and neurological deficit after exogenous TNF- was reversed by preinjection of 60 pmol mAb. Blocking endogenous TNF- also significantly reduced focal ischemic brain injury. Treatment with 60 pmol mAb before and after permanent MCAO significantly reduced infarct size compared with control (nonimmune) antibody treatment by 20.2% (P<.05). Reduced brain infarction also was produced by brain administration of 0.3 nmol (decreased 18.2%) or 0.7 nmol (decreased 26.1%; P<.05) sTNF-RI before and after focal stroke. The intracerebroventricular administration of TNF- or sTNF-RI did not alter brain or body temperature, blood gases or pH, blood pressure, blood glucose, or general blood chemistry. In cultured cerebellar granule cells, the application of TNF- did not directly affect neurotransmitter release or glutamate or oxygen free radical toxicity.

Conclusions These studies demonstrate that exogenous TNF- exacerbates focal ischemic injury and that blocking endogenous TNF- is neuroprotective. The specificity of the action(s) of TNF- was demonstrated by antagonism of its effects with specific anti–TNF- tools (ie, mAb and sTNF-RI). TNF- toxicity does not appear to be due to a direct effect on neurons or modulation of neuronal sensitivity to glutamate or oxygen radicals and apparently is mediated through nonneuronal cells. These data suggest that inhibiting TNF- may represent a novel pharmacological strategy to treat ischemic stroke.

Key Words: cerebral ischemia • middle cerebral artery occlusion • neuroprotection • tumor necrosis factor • rats

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Tumor necrosis factor- is a pleiotropic cytokine and appears to be involved in blood-brain barrier, inflammatory, thrombogenic, and vascular changes associated with brain injury (for reviews, see References 1 through 3^{1 2 3} ). TNF- levels in brain tissue, cerebrospinal fluid, and plasma have been found to be elevated in several central nervous system disorders, including Alzheimer's disease,⁴ multiple sclerosis,⁵ Parkinson's disease,⁶ meningococcal meningitis,⁷ and HIV infection.⁸ The major cellular elements capable of TNF- production in the brain have been identified in many different cell types after various types of stimulation/injury, including ependymal cells of the choroid plexus,⁹ activated astrocytes and microglia,^{10 11 12} and microglia and macrophages after ischemia^{13 14 15} as well as in central neurons after lipopolysaccharide treatment,¹⁶ cerebral focal ischemia,¹³ and brain injury.¹⁷

We previously demonstrated that early, increased neuronal expression of TNF- mRNA in the rat ischemic cortex preceded the leukocyte infiltration that occurs after focal stroke.^{13 18 19} Recently, early TNF- mRNA and protein expression were identified in microglia and activated macrophages after ischemia in the rat¹⁴ and mouse¹⁵ brain. The direct administration of TNF- into the brain produces a dramatic increase in leukocyte adhesion to vascular walls and an infiltration of these inflammatory cells into tissue but no direct neurotoxicity to neurons at the site of injection.¹³

Based on the available data, we hypothesized that applying TNF- directly to the brain would exacerbate the degree of injury after cerebral ischemia. In addition, to test the hypothesis that endogenous TNF- is an important mediator of focal ischemic injury, we used two different but specific methods of blocking TNF- during ischemia: an anti–TNF- monoclonal antibody (mAb) and a soluble TNF receptor. Finally, based on these earlier studies we hypothesized that the effects of TNF- on ischemic injury may not be due to its direct effects on neurons. Therefore, the aims of the present series of experiments were (1) to determine the effects of TNF-, preadministered into the brain cerebroventricular system, on the degree of brain injury and neurological deficits after focal stroke; (2) to determine the neuroprotective effects of blocking TNF- with the use of a specific antibody and a soluble receptor to TNF-; and (3) to study the direct effects of TNF- on cultured neurons to begin to evaluate potential mechanisms for TNF- in the mediation of ischemic brain injury.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Animals
Focal ischemia experiments were performed on male SHR (Taconic Farms, Germantown, NY) (weight range, 280 to 340 g). SHR were chosen because they exhibit a more consistent degree of damage after permanent or transient focal ischemia than normotensive rats.²⁰ Body temperature was maintained at 37°C during all surgical procedures and during recovery from anesthesia (ie, until normal locomotor activity was observed). For primary neuronal cell culture experiments, 8-day-old Sprague-Dawley rat pups (Taconic Farms) were used.

Animals were housed and cared for in accordance with the Guide for the Care and Use of Laboratory Animals (Bethesda, Md: Office of Science and Health Reports, DRR/NIH; 1985. US Dept of Health, Education, and Welfare [Dept of Health and Human Services] publication 85-23). Procedures in which laboratory animals were used were approved by the Institutional Animal Care and Use Committee of SmithKline Beecham Pharmaceuticals.

Effects of Brain TNF- Administration on Focal Ischemia
A study to evaluate the effects of brain TNF- pretreatment on focal stroke injury was executed during a 3-day period. On day 1 the animals were anesthetized with isoflurane (4% for induction and 2% for maintenance) and placed in a small-animal stereotaxic frame (David Kopf Instruments) for ICV injections. A sterile stainless steel cannula (OD=0.4 mm) was inserted into the left lateral ventricle through a small hole drilled in the skull (AP=0.2, L=1.5 from bregma, and V=3.2 from the dural surface with the skull level). A 5-µL solution of either TNF- (50 ng as 2.5 pmol or 500 ng as 25 pmol) or the vehicle (sterile PBS, containing 0.1% BSA) was infused into the ventricle with a microinfusion pump (Harvard Apparatus 22) over 20 minutes. The cannula was left in place for an additional 5 minutes to allow full absorption of the solution into the ventricle. In some groups 25 minutes before the left ventricular infusion of TNF- or the PBS vehicle, a 5-µL solution of anti–TNF- monoclonal antibody (mAb) (10 µg or 60 pmol), purified IgG antibody (IgG) (10 µg or 60 pmol as a nonimmune antibody control), or vehicle (PBS) was injected into the right ventricle. The recombinant mouse TNF- and monoclonal hamster anti-murine TNF- were obtained from Genzyme Diagnostics, and purified hamster IgG was obtained from Rockland Inc. On day 2 (24 hours after these ICV injections), the animals were anesthetized with pentobarbital (65 mg/kg IP) and underwent permanent MCAO for 24 hours or transient (80 or 160 minutes) MCAO followed by 24 hours of reperfusion, as described previously.^{19 20 21} Sham surgery was produced in some animals, also as described previously.^{18 19 20} On day 3 (24 hours after permanent MCAO or after 24 hours of reperfusion following transient MCAO), each rat was evaluated for neurological deficits with a graded scoring system, as previously described.²² In this scale, grade 1 denotes any amount of consistent contralateral forelimb flexion, grade 2 denotes grade 1 plus reduced resistance to lateral shift toward the contralateral side, and grade 3 denotes grade 2 deficits plus circling behavior toward the paretic side. Grade 0 was assigned to animals with no consistent deficits. Rats were then killed by an overdose of sodium pentobarbital (100 mg/kg IP). The brains were immediately removed, and 2-mm coronal sections were cut from the entire forebrain area (ie, from the olfactory bulbs to the cortical-cerebellar junction) with a brain slicer (Zivic-Miller Laboratories). The coronal sections were immediately stained in a solution of 1% triphenyltetrazolium chloride, as described previously.²³ Sections were transferred to 10% formalin (in 0.1% sodium phosphate buffer) for at least 24 hours and then photographed and analyzed with the use of an image analysis system, as described previously.^{20 21} Infarct size was expressed as the percent infarcted tissue in reference to the contralateral hemisphere, and hemispheric swelling was expressed as the percent increase in the size of the ipsilateral hemisphere compared with the contralateral hemisphere. Infarct volume (in cubic millimeters) was also determined for each animal.

Effects of Blocking Brain TNF- on Focal Ischemia
Endogenous (central) TNF- was blocked before and during focal stroke by repeated ICV administrations of mAb or sTNF-RI. These studies were conducted over a 3-day period. On day 1, animals were anesthetized with pentobarbital, and a stainless steel guide cannula (OD=1.0 mm) was cemented in place at the same stereotaxic coordinates as described above, except that D=2.7 from the dural surface (ie, located just over but not penetrating the right lateral ventricle). On day 2, 5 µL containing appropriate control solution/vehicles, 10 µg (60 pmol) mAb, and 5 µg (0.3 nmol) or 12.5 µg (0.7 nmol) sTNF-RI was administered over a 2-minute period at 30 minutes before and 3 and 6 hours after permanent MCAO with the use of a cannula (OD=0.4 mm) that just penetrated the ventricle when positioned into the guide cannula. Control solution for mAb was 10 µg (60 pmol) nonimmune IgG. The vehicle for sTNF-RI was PBS containing 0.1% BSA. The sources for mAb and IgG were as described above. The source for recombinant human sTNF-RI was R&D Systems. On day 3 (24 hours after permanent MCAO), each rat was evaluated for neurological deficits and then killed, the brain was removed, and the forebrain was stained and analyzed as described above.

TNF- and sTNF-RI Effects on Physiological Measures
To determine whether more nonspecific central effects of TNF- or its blockade were associated with observed changes in brain injury, we evaluated the effects of ICV TNF- and sTNF-RI on several relevant physiological measures. SHR were instrumented for ICV administration (as described above) and for conscious recording of blood pressure, as described previously.²⁰ Rats were then anesthetized and received 25 pmol TNF-, 0.7 nmol sTNF-RI, or vehicle (PBS) as described above. Temporalis muscle (as an indirect index of brain temperature²⁴ ) and rectal (body) temperatures were measured as described previously²⁵ before ICV treatment and during 3 hours after ICV treatment (under anesthesia) and 24 hours later (in conscious animals). In addition, heparinized blood samples and mean arterial blood pressure data (averaged from 5-minute intervals) were collected from each of the instrumented animals over similar time periods. Blood samples were evaluated with an automated blood gas analyzer (Radiometry) for PO_2, PCO₂, and pH. Blood glucose and general blood chemistry (including sodium, potassium, calcium, albumin, creatinine, and liver enzymes) were also determined (Monarch 2000; Instrumentation Laboratory).

TNF- Effects on Cerebellar Neurons
Viability, transmitter release, and glutamate and oxygen free radical toxicity were evaluated in primary cultures of rat cerebellar neurons²⁶ prepared in groups of 20 cerebellums from the rat pups and used after 8 to 9 days in culture. Mixed cultures with astrocytes were obtained by withholding the mitotic inhibitor cytosine ß-D-arabinofuranoside. Cells were washed and incubated in a buffer composed of the following (mmol/L): NaCl 154, KCl 5.6, CaCl₂ 2.3, MgCl₂ 1.0, and HEPES 8.6 adjusted to pH 7.4 with NaOH. We have previously characterized the pharmacology of the N-methyl-D-aspartate response in these neurons, which offer a glutamate excitotoxicity model that is dependent on compromised energy levels and relief of the voltage-dependent Mg²⁺ block.^{26 27} Neurons (3x10⁶) in 35-mm dishes were incubated at 37°C in the above buffer for 40 minutes before the addition of glutamate. The process consisted of a wash, 10-minute preincubation, and 30-minute incubation, each in 1 mL of fresh buffer. Recombinant mouse TNF- and IL-1ß (Genzyme Diagnostics) were added in increasing concentrations to the growth medium as a sensitizing pretreatment, 6 or 24 hours before the experiments, and were subsequently removed with the washes. Thirty minutes after glutamate addition, cells were assessed for viability by staining with fluorescein diacetate (Sigma), which had been freshly diluted 1:1000 into incubation buffer from a stock of 10 mg/mL acetone. The staining solution (1 mL) was aspirated after 5 minutes and replaced with fresh buffer containing glucose for immediate counting of viable neurons by fluorescence microscopy. Viability was expressed as a percentage of the total number of cells retaining fluorescein. Toxicity data are averaged duplicate determinations from separate titrations from three different neuronal preparations that were prepared on separate days.

We also determined the effect of TNF- on delayed glutamate neurotoxicity 24 hours after glutamate administration (ie, a time that more closely mimics the in vivo time course associated with ischemic stroke toxicity) and following the procedure of Manev et el.²⁸ Neurons were treated overnight with up to 50 000 U/mL TNF-, the growth medium was removed, and cells were washed in the same buffer as above but containing 5.6 mmol/L glucose and lacking MgCl₂. Neurons were then exposed to increasing concentrations (10 µmol/L to 1 mmol/L) of glutamate for 5 minutes at room temperature and washed once in fresh complete buffer, the growth medium containing TNF- was replaced, and neurons were scored for toxicity after 24 hours.

For neurotransmitter release experiments, neurons in 35-mm dishes were washed and incubated in 1 mL buffer with 1 µCi D-[2,3-³H]aspartic acid (NET-581, DuPont NEN Research Products) for 15 minutes at 37°C, as previously described.^{29 30} Labeled neurons were rapidly washed twice in buffer and incubated with 1 mL buffer that was sequentially removed and replaced every 4 minutes (4 times over 16 minutes). Veratridine or KCl was then added to induce efflux and maintained throughout for 24 minutes, with six replacements at 4-minute intervals. Experiments were terminated by adding 500 µL (twice) of 0.1% NaOH to solubilize cells, and radioactivity was determined along with the buffer samples by liquid scintillation counting in 10 mL Ready Safe (Beckman Instruments). Rate constants (k) for [³H]aspartate efflux at each time interval were calculated according to the following equation:

where A₁ and A₂ represent the total counts remaining in the dish at times t₁ and t₂, respectively. Data were also calculated as fractional loss of total disintegrations per minute released to calculate percent total release after veratridine or KCl stimulation. TNF- and IL-1ß were added in increasing concentrations to the growth medium in separate dishes 6 or 24 hours before the [³H]aspartate efflux experiments.

For the oxygen free radical experiments, cerebellar granule cell cultures grown in 35-mm dishes were deprived of glucose for 40 minutes before being exposed to a free radical–generating system (DHF-Fe³⁺/ADP) composed of final concentrations (mmol/L) of DHF 0.83, FeCl₃ 0.025, and ADP 0.25, which generates superoxide anions and hydroxyl radicals.³¹ SB 211475 is a potent metabolite of carvedilol and was added to neurons in the glucose-free buffer described above 20 minutes before addition of the free radical–generating system, as described previously.³² TNF- was added to the growth medium 6 hours before the experiments. After neurons were exposed to the DHF-Fe³⁺/ADP for 20 minutes, the free radical–generating system was removed and replaced with glucose-free buffer. Toxicity was assessed 50 to 60 minutes later by staining with fluorescein diacetate, as previously described.^{31 32} Protection of neurons was determined by counting cells and expressing the percent viability.

Statistical Analysis
Results are presented as mean±SE. Statistical analyses of all data were performed with the use of ANOVA with Dunnett or Tukey follow-up tests or t test if appropriate. Significance was accepted at P<.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
TNF- Exacerbates Stroke Injury
Animals injected ICV with vehicle 24 hours before MCAO exhibited a 13.1±1.3% hemispheric infarct. Injection of 2.5 pmol TNF- 24 hours before the permanent MCAO produced a significant (P<.05) increase in the infarct size to 18.9±1.7%. Injection of 25 pmol TNF- produced an even greater (P<.01) increase in the infarct size to 27.1±1.3% (Fig 1A). In vehicle-injected animals there was a 4.6±0.7% increase in the ipsilateral hemispheric size (swelling). In the 2.5-pmol TNF- group there was a significant (P<.05) augmentation in swelling (up to 9.8%±1.6%) compared with the vehicle-injected animals. There was no significant increase in swelling in the group injected with 25 pmol TNF-. The neurological score for the sham-operated animals was 0±0 (n=4). In the vehicle-injected group the neurological score was 1.6±0.2. There was a significant increase in the score for the 25-pmol TNF- group (up to 2.3±0.2; P<.05) and a small, nonsignificant increase in the 2.5-pmol TNF- group (up to 1.7±0.2) (Fig 1B).

View larger version (35K): [in this window] [in a new window]   Figure 1. TNF- increases ischemic injury and neurological deficits produced by permanent MCAO. A, Dose-related increase in percent hemispheric infarct is produced by 2.5 and 25 pmol ICV TNF-. B, TNF- (25 pmol ICV) increases neurological grade. n=7 to 12 per bar group. *P<.05 vs PBS vehicle, **P<.05 vs PBS vehicle and 2.5 pmol TNF- (ANOVA with Tukey test follow-up).

In the animals undergoing 80 minutes of MCAO with reperfusion, 2.5 pmol TNF- produced a significant (P<.05) increase in infarct size (4.3±0.4%) compared with the animals receiving vehicle (1.9±0.9%) (Fig 2). There was no significant increase in swelling in this group of animals, and only small variable neurological deficits were observed (data not shown). In the animals undergoing 160 minutes of MCAO with reperfusion, 2.5 pmol TNF- also produced a significant (P<.05) increase in infarct size (21.6±2.2%) compared with the animals receiving vehicle (14.2±1.3%) (Fig 2). There was no significant difference in swelling between the groups receiving vehicle (3.2±1.7%) or 2.5 pmol TNF- (2.7±0.8). There was a small, nonsignificant increase in the neurological score in the 2.5-pmol TNF- group (1.2±0.4) compared with the vehicle group (0.8±0.2).

View larger version (34K): [in this window] [in a new window]   Figure 2. Increasing duration of transient MCAO (from 80 to 160 minutes) followed by 24 hours of reperfusion produces increasing degrees of percent hemispheric infarct (P<.05). TNF- (2.5 pmol ICV) significantly increases the degree of ischemic injury above that of the PBS vehicle control treatment (P<.05). n=5 to 7 per bar group. *P<.05 vs PBS vehicle (ANOVA with Dunnett test follow-up).

Vehicle-injected (ie, PBS into both right and left ventricles) rats that received permanent MCAO exhibited a 19.5±0.9% hemispheric infarct. Injection of exogenous (25 pmol) TNF- into the left ventricle (after PBS vehicle injection into the right ventricle) significantly (P<.05) increased infarct size to 29.2±0.9% (Fig 3A). Preinjection of 60 pmol TNF- mAb into the right ventricle 30 minutes before exogenous (25 pmol) TNF- into the left ventricle significantly (P<.0001) reduced the exacerbation observed with this dose of TNF- (ie, decreased to 21.5±0.9%). There was no significant difference in infarct size between the animals receiving mAb plus PBS injections compared with animals receiving bilateral PBS injections (Fig 3A), indicating that a single pretreatment with mAb was not sufficient to reduce ischemic injury due to endogenous TNF-. The percent hemispheric infarct in the animals receiving 60 pmol purified hamster IgG in the right ventricle plus 25 pmol TNF- in the left ventricle (29.3±3.7%) was not significantly different from that in the group receiving PBS in the right ventricle plus 25 pmol TNF- in the left ventricle.

View larger version (39K): [in this window] [in a new window]   Figure 3. TNF- (25 pmol) exacerbation of hemispheric infarction (A), hemispheric swelling (B), and neurological grade (C) is reversed by mAb (10 µg or 60 pmol) but not by purified control antibody (IgG; 10 µg or 60 pmol). R/L indicate 5-µL treatments into right and left lateral ventricles administered 24 hours before permanent MCAO, as described in text. n=4 to 7 per bar group. *P<.05 vs R-PBS/L TNF-; sham is different from all other groups (P<.05) (ANOVA with Tukey test follow-up).

Permanent MCAO in bilateral PBS-injected animals produced a 3.8±1.4% increase in hemispheric swelling (Fig 3B). Injection of 25 pmol TNF- into the left ventricle (before PBS injection into the right ventricle) significantly (P<.05) increased swelling to 9.2±0.9%. Preinjection of 60 pmol mAb into the right ventricle 30 minutes before 25 pmol TNF- into the left ventricle significantly (P<.05) reduced the exacerbation in swelling observed with 25 pmol TNF- (ie, down to 2.2±0.7%). There was no significant difference in tissue swelling between animals receiving mAb plus PBS, animals receiving mAb plus TNF-, or animals receiving bilateral PBS injections. Hemispheric swelling in the animals receiving nonimmune IgG into the right ventricle plus TNF- into the left ventricle (9.7±1.9%) was not significantly different from the group receiving PBS into the right ventricle plus 25 pmol TNF- into the left ventricle. In all studies, changes in infarct volume were identical to those that occurred in percent hemispheric infarct size (data not shown).

Neurological grade for the rats injected bilaterally with PBS undergoing permanent MCAO was 1.6±0.2 (Fig 3C). Injection of TNF- into the left ventricle (before PBS injection into the right ventricle) significantly (P<.05) increased the neurological grade to 2.1±0.1. Preinjection of 1.7 nmol mAb into the right ventricle 30 minutes before TNF- into the left ventricle significantly (P<.05) reduced the worsening of the neurological score observed with TNF- (ie, decreased down to 1.5±0.2). There was no significant difference in the scores between animals receiving TNF- plus PBS, animals receiving mAb plus TNF-, or animals receiving bilateral PBS injections. The score in the animals receiving nonimmune IgG plus TNF- was not significantly different from the group receiving PBS plus TNF- injections.

Blocking TNF- Reduces Stroke Injury
Based on earlier demonstrations of increased central neuronal and microglial expression of TNF- after focal stroke,^{13 14 15} endogenous (central) TNF- was blocked before and during focal stroke by repeated ICV administrations of mAb or sTNF-RI. Treatment before and during 24 hours of focal ischemia with repeated ICV mAb or sTNF-RI significantly reduced focal ischemic injury. Fig 4A illustrates the results on percent hemispheric infarct for 60 pmol mAb compared with control (equivalent amount of nonimmune IgG) treatment. Identical results were also observed on infarct volume that also was measured in the same animals (control=147.2±7.5 mm³ compared with mAb=117.4±7.0; P<.05). Fig 4B depicts the results on percent hemispheric infarct for sTNF-RI compared with vehicle (PBS plus 0.1% BSA) treatments. Again, identical results were obtained on infarct volume measured in the same animals (vehicle=119.5±7.5 mm³; 5 µg or 0.3 nmol sTNF-RI=97.8±8.5 mm³; 12.5 µg or 0.7 nmol sTNF-RI=89.0±7.6 mm³; P<.05). The small degree of percent hemispheric swelling that occurred under these conditions (2.3±0.5% to 1.9±0.8%) was not altered by the TNF-–blocking treatments. In addition, neurological scores (1.9±0.1% to 1.8±0.2%) were not altered by these treatments.

View larger version (24K): [in this window] [in a new window]   Figure 4. A, mAb (10 µg or 60 pmol; 3x) significantly reduced focal ischemic injury compared with purified control antibody treatment (IgG; 10 µg or 60 pmol). n=5 to 7 per bar group. *P<.05 (unpaired t test). B, sTNF-RI (5 µg; 0.3 nmol or 12.5 µg; 0.7 nmol) significantly reduced focal ischemic injury compared with vehicle (PBS) treatment. n=8 to 14 per bar group. *P<.05 (ANOVA with Dunnett test follow-up).

Brain TNF- Manipulation Did Not Affect Other Monitored Physiological Variables
To evaluate potential nonspecific effects of ICV treatments that may have influenced brain injury, several relevant variables were monitored. Table 1 lists the effects of ICV vehicle (PBS), TNF-, and sTNF-RI on temporalis muscle and rectal temperatures. Table 2 lists the effects of these ICV treatments on blood pressure. Table 3 lists these results for blood gases, pH, and glucose. No significant differences for any measures were observed between ICV treatments. In addition, no differences due to ICV treatments were observed for any of the other measurements of blood chemistry (data not shown).

View this table: [in this window] [in a new window]   Table 1. Effects of ICV Administration of Vehicle (PBS), TNF-, or sTNF-RI on Temporalis (Brain) and Rectal (Body) Temperature

View this table: [in this window] [in a new window]   Table 2. Effects of ICV Administration of Vehicle (PBS), TNF-, or sTNF-RI on Mean Arterial Blood Pressure

View this table: [in this window] [in a new window]   Table 3. Effects of ICV Administration of Vehicle (PBS), TNF-, or sTNF-RI on Blood Gases, pH, and Glucose

TNF- Effects on Cerebellar Neurons
To determine whether the increase in ischemia-induced brain damage in rats pretreated with TNF- was due to direct effects of TNF- on neurons, we studied the effects of TNF- on cultured rat cerebellar granule cells grown in the presence or absence of astrocytes. Isolated neuronal cultures contained 90% to 95% neurons only. Mixed neuronal cultures contained neurons grown above a monolayer of astrocytes, similar to that described for cortical cultures. Pretreatment for 24 hours with increasing concentrations of TNF- did not result in increased release of neurotransmitter from cultured neurons. This was true for [³H]aspartate efflux induced with either 50 mmol/L KCl (Fig 5A) or 40 µmol/L veratridine (Fig 5B). Since IL-1ß is also upregulated in ischemic stroke,^{18 33} it could act in concert with TNF- to sensitize neurons to TNF- actions. Therefore, in a series of additional experiments, IL-1ß was included at concentrations up to 48 ng/mL, with 100 U/mL TNF- as a 24-hour pretreatment. This also did not result in increased release of [³H]aspartate efflux induced with either 50 mmol/L KCl (100.4±8.1%) or 40 µmol/L veratridine (94.1±4.9%) (n=7 for each condition).

View larger version (28K): [in this window] [in a new window]   Figure 5. TNF- does not result in increased release of neurotransmitter from cultured neurons. TNF- was added to neuronal cultures at the indicated concentrations for 24 hours before the experiments. A, Results from [3H]aspartate efflux experiments are expressed as the percent total release achieved by 50 mmol/L KCl (n=8). B, Results are compared with the percent total release achieved by 40 µmol/L veratridine (n=4).

We also studied the effects of TNF- and IL-1ß on glutamate neurotoxicity in cultured neurons. As seen in Fig 6, 24-hour pretreatment with up to 2000 U/mL TNF- and 100 ng/mL IL-1ß did not shift the dose response for glutamate excitotoxicity compared with neurons receiving only glutamate. Also, TNF- or IL-1ß alone had absolutely no toxic effects on the neurons (ie, >95% viability in eight separate experiments). Because the above experiments are a model of acute glutamate toxicity (death in 30 minutes), while MCAO results are assessed after 24 hours, we also studied glutamate toxicity in cultured neurons after a delay period (ie, in a 24-hour model²⁸ ). Pretreatment with from 5000 to 50 000 U/mL TNF- did not directly injure the cultured neurons, nor did it significantly shift the dose-response curve for delayed glutamate toxicity (n=3; data not shown). Therefore, TNF- pretreatment had no direct effect on either acute or delayed glutamate excitotoxicity in this well-characterized model system.

View larger version (24K): [in this window] [in a new window]   Figure 6. TNF- in combination with IL-1ß does not enhance glutamate excitotoxicity. TNF- (units per milliliter) and IL-1ß (nanograms per milliliter) were added to neuronal cultures at the indicated concentrations for 24 hours before the experiments. Data are averaged duplicate determinations from separate titrations of glutamate alone (n=10) or with cytokine pretreatment (n=4 each). Results are expressed as percent toxicity compared with untreated control cultures.

To investigate whether TNF- might act on neurons at an earlier time point, we performed similar tests after only 6 hours of incubation with 1000 U/mL TNF-. We also compared the effects of TNF- on neurons grown in the presence of astrocytes, which might have an influence on neuronal sensitivity/toxicity. As shown in Fig 7A and 7D, there was no increase in sensitivity to glutamate excitotoxicity with TNF- pretreatment, although higher glutamate concentrations were necessary to kill neurons in mixed culture. Incubation with TNF- for 6 hours also did not alter the veratridine-induced release of [³H]aspartate (Fig 7B and 7E). We also tested the ability of TNF- to alter the protective effects of a free radical scavenger and thus influence cell damage by an oxygen free radical–generating system. We have previously shown that neuronal cell death was complete 50 minutes after a 20-minute exposure to a DHF-Fe³⁺/ADP oxygen free radical–generating system.^{31 32} Here the inclusion of 1000 U/mL TNF- as a 6-hour pretreatment did not change the protective dose-response curve for the free radical scavenger SB 211475, which had an EC₅₀ of approximately 300 nmol/L with or without TNF- pretreatment (Fig 7C and 7F).

View larger version (42K): [in this window] [in a new window]   Figure 7. Short-term treatment with TNF- does not influence neuronal responses to a variety of insults. TNF- was added to cerebellar granule cells in either pure (A, B, C) or mixed (D, E, F) cultures at a concentration (CONC) of 1000 U/mL 6 hours before the following experiments: A and D, glutamate excitotoxicity; B and E, veratridine-induced [3H]aspartate release; C and F, neurotoxicity induced with a DHF-Fe3+/ADP free radical–generating system (FRGS). Results are expressed as mean±SEM of three replicates for each set of experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Pretreatment of the brain with exogenous TNF- exacerbated infarct size in both transient and permanent focal stroke models but did not produce nonspecific effects on brain or body temperature, blood pressure, blood gases or pH, or blood chemistry. The ICV administration of TNF- was expected to provide brain concentrations of the cytokine similar to the elevated pathological levels previously measured in rodents and humans.^{34 35} A TNF- dose-related effect was demonstrated after permanent MCAO (2.5 and 25 pmol), and a consistent increase in injury was observed after MCAO with reperfusion (80 minutes and/or 160 minutes) with 2.5 pmol TNF. The effect of TNF- was demonstrated to be specific by the ability of mAb to completely reverse the exacerbation of infarct size, swelling, and neurological deficit, while the nonimmune IgG antibody control treatment had no such effect. It is interesting that pretreatment with a single ICV administration of mAb was not adequate to reduce ischemic brain injury (ie, that might be attributed to endogenous TNF-), but this was effectively reduced by repetitive administrations before and during ischemia (see below).

The monoclonal hamster anti-murine TNF- effectively neutralizes the bioactivity of TNF-, has been shown to be specific for (ie, effectively antagonizes) both mouse and rat TNF-,^{36 37} and has been demonstrated previously to prevent the transfer of experimental allergic encephalomyelitis.³⁸ Based on these studies, the ICV mAb dose used in the present studies was expected to provide a brain concentration of the antibody that would block TNF-–mediated effects. Indeed, repeated ICV administrations of mAb just before and during focal ischemia produced a significant reduction in ischemic brain injury.

Soluble TNF receptors are truncated forms of the extracellular domains of the receptors and act as endogenous inhibitors of TNF- by competing with the cell-surface receptors for its binding.^{39 40} sTNF-RI has already demonstrated efficacy in other brain injury models that are associated with elevated TNF- levels (eg, brain trauma,⁴¹ experimental autoimmune encephalomyelitis,^{42 43} and lethal endotoxemia³⁵ ). For example, similar to focal stroke, brain TNF- levels are increased after head injury,^{44 45 46} and inhibition of TNF- with sTNF-RI provides cerebroprotection.⁴¹ In the present studies, the repeated ICV doses of sTNF-RI were used to provide brain concentrations of the soluble receptor that were expected to block TNF- (ie, similar to that achieved in other models in which efficacy was demonstrated^{35 41 42 43} ). These brain administrations of sTNF-RI produced a significant decrease in focal stroke injury. In addition, another laboratory has simultaneously described neuroprotection due to sTNF-RI administration in focal stroke. Intravenous sTNF-RI significantly reduces the impairment in cortical microvascular perfusion that accompanies MCAO and also decreases the degree of brain injury after MCAO, strongly suggesting a vascular mechanism for TNF- in stroke.⁴⁷ In addition, direct application of sTNF-RI to the ischemic cortex significantly reduces brain injury in murine MCAO (J.M. Hallenbeck, MD, personal communication, 1996), thus corroborating the present results. All these new data support previous reports on TNF- mRNA and peptide expression in the same stroke model and provide a very strong case for a role of TNF- as a mediator of ischemic brain injury.

Hemispheric swelling was evaluated independently for determination of TNF- treatment effects. Exogenous TNF- pretreatment was able to exacerbate the small degree of swelling (an index of edema²⁰ ) that occurs under these conditions. However, blocking endogenous TNF-, although able to significantly reduce brain infarction, did not alter swelling. Effects on infarct size independent of swelling in focal stroke have been observed previously^{48 49} and indicate that neuroprotection provided by blocking endogenous TNF- in the present studies is not related to the effects of altered swelling and its influence on the degree of infarction. In addition, although increased neurological scores related to significantly increased brain infarctions were observed as a result of TNF- pretreatment, neurological deficits were not reduced by blocking TNF-, as might be expected from previous data indicating that larger reductions in infarct size are necessary to affect this less sensitive measure.⁴⁸

Although many studies were conducted, we could provide no direct evidence for TNF- toxicity on neurons in relatively pure or mixed cultures. Excess glutamate release is known to occur in ischemic stroke, and the present studies were conducted with or without glucose in the presence of glutamate (ie, both acute and delayed standard models of neurotoxicity). In addition, TNF- together with IL-1ß did not shift the dose response for glutamate neurotoxicity in these cultured neurons. No augmented release of [³H]aspartate from the cultured neurons was evident when TNF- with or without IL-1ß was present in the culture even at very high concentrations. Similarly, TNF- pretreatment did not enhance the toxic effects of a free radical–generating system, nor did it shift the dose-response curve of a protective antioxidant. Therefore, an indirect augmentation of neuronal damage by TNF- seems likely. Astrocytes and microglia are the prime brain cell candidates likely to respond to cytokine stimulation and maintain an inflammatory response that will ultimately result in long-term neuronal loss after ischemic stroke. Evidence supporting this interpretation was recently provided by Chao and Hu,⁵⁰ using a model system of human fetal neurons cocultivated with astrocytes, when cultures were exposed to glutamate in the presence of TNF- for 6 days. Interestingly, IL-1, IL-1ß, transforming growth factor-ß1, and IL-6 had no effect on glutamate toxicity, while TNF- augmented glutamate toxicity indirectly by altering glutamate metabolism in the astrocyte component of the coculture. TNF- decreased astrocytic glutamine synthetase activity by 27%, as previously shown in cultured murine astrocytes,⁵¹ while also inhibiting high-affinity glutamate uptake by 50%.⁵⁰ It is likely that these augmented effects were due to gene induction over time, since TNF- does induce a p53-dependent apoptosis in rat glioma cells.⁵² Such a disruption in the integrity of support cells could certainly contribute to increased infarction after MCAO. Our present results, showing no direct, acute effect of TNF- on pure or mixed neuronal cultures, are limited in that they are a model system and not identical to the in vivo situation. However, they do support the possibility that the effects of TNF- in focal stroke might be due to an augmented toxicity induced by other cells in the brain after ischemia. These data are consistent with our earlier work that illustrated no obvious neuronal damage when TNF- was injected into the cortex but did show an increase in leukocyte accumulation in blood vessels at the site of injection.¹³

Others have also shown that TNF- is not directly toxic to neurons,^{53 54} and some investigators even suggest a protective effect of TNF- on neurons.^{55 56 57} The broad scope of the injurious and beneficial effects of TNF- has been emphasized previously.⁵⁸ Cytokines have also been suggested to provide beneficial effects in brain injury, as inferred from studies with multiple-TNF-receptor knockout mice (p55 and p75 knockout) that display increased sensitivity to brain ischemia^{59 60} and the capacity of IL-1 to elicit a state of ischemic tolerance on repeated administration.⁶¹ We have demonstrated previously the later expression of TNF- in macrophages involved in resolution of ischemic brain injury.^{13 62} Certainly the eventual determination of the role of individual known and novel TNF receptors will help in our understanding of how to best target TNF- in tissue injury. Blocking TNF effects, especially at TNF receptor I, might be expected to block not onlyTNF-mediated apoptosis but also TNF activation of nuclear factor-B (NF-B)–associated inflammation.⁶³ However, the present data clearly demonstrate that increasing the acute effects of TNF- is not protective (and in fact increases ischemic injury) and that blocking the acute increased activity of TNF- that occurs after focal stroke is neuroprotective.

The toxic effects of TNF- and its role as a mediator of focal ischemia may involve many mechanisms. For example, TNF- increases blood-brain barrier permeability and produces pial artery constriction^{64 65} that can contribute to focal ischemic brain injury, and there appears to be a direct toxic effect of TNF- on the capillary.^{66 67} TNF- increases capillary permeability and opens the blood-brain barrier, apparently by increasing matrix-damaging metalloproteinase (gelatinase B) production,⁶⁸ which is also expressed early after focal stroke.⁶⁹ TNF- also causes damage to myelin and oligodendrocytes^{70 71} and increases astrocytic proliferation,⁷² thus potentially contributing to demyelination and reactive gliosis during brain injury. In addition, TNF- activates the endothelium for leukocyte adherence and procoagulation activity (ie, increased tissue factor, von Willebrand factor, and platelet activating factor) that can exacerbate ischemic damage.⁷³ Indeed, increased TNF- in the brain and blood in response to lipopolysaccharide appears to contribute to increased brain stem thrombosis and hemorrhage^{74 75 76} and can contribute to increased stroke sensitivity/risk in hypertensive rats.²⁰ Finally, TNF- plays a pivotal role in inflammatory processes.⁷⁷ It activates neutrophils,⁷⁸ increases leukocyte–endothelial cell adhesion molecule expression,⁷³ and increases leukocyte adherence to blood vessels and their subsequent infiltration into the brain.¹³ Clearly, leukocyte transit through capillaries is impaired after stroke, which contributes to rheological effects due to microvascular occlusion or plugging.^{79 80} The efficacy of neutrophil depletion and the blocking of leukocyte-associated adhesion molecules has been demonstrated repeatedly (for reviews, see References 1 and 81^{1 81} ). TNF- also stimulates the release of potent vasoactive agents⁸² and induces both vasodilation⁸³ and vasoconstriction and reduced cerebral blood flow,⁸⁴ thus modifying blood flow to already ischemic tissue and perhaps increasing the potential of leukocyte plugging.

The present studies indicate that increasing brain TNF- before focal stroke can be expected to exacerbate ischemic injury and that blocking TNF- in a stroke can be expected to reduce focal ischemic injury. These changes in the degree of ischemic brain injury do not appear to be related to nonspecific effects of altering brain TNF-. Clearly, more studies are necessary to understand the mechanisms for TNF- mediation of ischemic injury. However, the available data suggest that blocking this cytokine should be an attractive pharmacological goal in reducing ischemic injury. Similar to interference with IL-1 receptors, which reduces brain injury after focal ischemia^{85 86} and head injury,⁸⁷ blocking TNF- has now proven to be protective in focal stroke (present studies and Reference 47⁴⁷ ) and head trauma.³⁷ The evaluation of additional potent and specific anti-TNF therapeutics (see Reference 2² for review) in proper models of stroke is clearly warranted.

	Selected Abbreviations and Acronyms

 

BSA = bovine serum albumin DHF = dihydroxyfumarate ICV = intracerebroventricular, intracerebroventricularly IgG = IgG antibody IL = interleukin mAb = anti–TNF- monoclonal antibody MCAO = middle cerebral artery occlusion SHR = spontaneously hypertensive rats sTNF-RI = soluble TNF receptor I TNF = tumor necrosis factor

	Acknowledgments

 
The authors thank Shirley Wilson and Sue Tirri for assistance in the preparation of this manuscript and Kathy Morasco for performing blood chemistry determinations in the present studies.

	Footnotes

 
Presented in part at the 21st International Joint Conference on Stroke and Cerebral Circulation, San Antonio, Tex, January 25-27, 1996, and published in abstract form (Stroke. 1996;27:187).

Received September 11, 1996; revision received February 5, 1997; accepted March 12, 1997.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Feuerstein GZ, Liu T, Barone FC. Cytokines, inflammation and brain injury: the role of tumor necrosis factor . Cerebrovasc Brain Metab Rev. 1994;6:341-360.[Medline] [Order article via Infotrieve]
2. Arvin B, Neville LF, Barone FC, Feuerstein GZ. The role of inflammation and cytokines in brain injury. Neurosci Biobehav Rev. 1996;20:445-452.[Medline] [Order article via Infotrieve]
3. Kochanek PM, Hallenbeck JM. Polymorphonuclear leukocytes and monocytes/macrophages in the pathogenesis of cerebral ischemia and stroke. Stroke. 1992;23:1367-1379.[Abstract/Free Full Text]
4. Fillit H, Ding W, Buee L, Kalman J, Altstiel L, Lawlor B, Wolf-Klein G. Elevated circulating tumor necrosis factor levels in Alzheimer's disease. Neurosci Lett. 1991;131:318-320.
5. Sharief MK, Noori MA, Ciardi M, Cirelli A, Thompson EJ. Increased levels of circulating ICAM-1 in serum and cerebrospinal fluid of patients with active multiple sclerosis: correlation with TNF and blood brain barrier damage. J Neuroimmunol. 1993;43:15-22.[Medline] [Order article via Infotrieve]
6. Mogi M, Harada M, Riederer P, Narabayashi H, Fujita K, Nagatsu T. Tumor necrosis factor- (TNF-) increases both in the brain and in the cerebrospinal fluid from parkinsonian patients. Neurosci Lett. 1994;165:208-210.[Medline] [Order article via Infotrieve]
7. Waage A, Holstensen A, Shalaby R, Brandtzaeg P, Kierulf P, Espevick T. Local production of tumor necrosis factor , interleukin-1 and interleukin-6 in meningococcal meningitis: relation to the inflammatory response. J Exp Med. 1989;170:1859-1867.[Abstract/Free Full Text]
8. Merrill JE, Chen ISY. HIV-1, macrophages, glial cells, and cytokines in AIDS nervous system disease. FASEB J. 1991;5:2391-2397.[Abstract]
9. Tarlow MJ, Jenkins R, Cosmis SD, Osborne MP, Stephens S, Stanley P, Crocker J. Ependymal cells of the choroid plexus express tumor necrosis factor-. Neuropathol Appl Neurobiol. 1993;19:324-328.[Medline] [Order article via Infotrieve]
10. Chung IY, Benveniste EN. TNF production by astrocytes: induction by lipopolysaccharide, interferon-gamma and interleukin-1. J Immunol. 1990;144:2999-3007.[Abstract]
11. Liberman AP, Pitha PM, Shin HS, Shin ML. Production of tumor necrosis factor and other cytokines by astrocytes stimulated with lipopolysaccharide or a neurotropic virus. Proc Natl Acad Sci U S A.. 1989;86:6348-6352.[Abstract/Free Full Text]
12. Ganter S, Northhoff H, Männel D, Gibiche-Haerter PJ. Growth control of cultured microglia. J Neurosci Res. 1992;33:218-230.[Medline] [Order article via Infotrieve]
13. Liu T, Clark RK, McDonnell PC, Young PR, White RF, Barone FC, Feuerstein GZ. Tumor necrosis factor- expression in ischemic neurons. Stroke. 1994;25:1481-1488.[Abstract]
14. Buttini M, Appel K, Sauter A, Gebicke-Haerter P-J, Boddeke HWGM. Expression of tumor necrosis factor alpha after focal cerebral ischemia in the rat. Neuroscience. 1996;71:1-16.[Medline] [Order article via Infotrieve]
15. Uno H, Matsuyama T, Tagaya M, Hata R, Akita H, Nakamura H, Sugita M. Cerebral ischemia triggers early production of tumor necrosis factor- by microglia. J Cereb Blood Flow Metab. 1995;15(suppl 1):118. Abstract.
16. Breder CD, Hazulea C, Ghayur T, Klug C, Huginin M, Yasuda K, Teng M, Saper CB. Regional induction of tumor necrosis factor expression in the mouse brain after systemic lipopolysaccharide administration. Proc Natl Acad Sci U S A.. 1994;91:11393-11397.[Abstract/Free Full Text]
17. Tchelingerian JL, Quinonero J, Booss J, Jacque C. Localization of TNF and IL-1 immunoreactivities in striatal neurons after surgical injury to the hippocampus. Neuron. 1993;10:213-224.[Medline] [Order article via Infotrieve]
18. Wang XK, Yue T-L, Barone FC, White RF, Young PR, McDonnell PC, Feuerstein GZ. Concomitant cortical expression of TNF and IL-1ß mRNA following transient focal ischemia. Mol Chem Neuropathol. 1994;23:103-114.[Medline] [Order article via Infotrieve]
19. Barone FC, Hillegrass LM, Tzimas MN, Schmidt DB, Foley JJ, White RF, Price WJ, Feuerstein GZ, Clark RK, Griswold DE, Sarau HM. Time-related changes in myeloperoxidase activity and leukotriene B₄ receptor binding reflect leukocyte influx in cerebral focal stroke. Mol Chem Neuropathol. 1995;24:13-30.[Medline] [Order article via Infotrieve]
20. Barone FC, Price WJ, White RF, Willette RN, Feuerstein GZ. Genetic hypertension and increased susceptibility to cerebral ischemia. Neurosci Biobehav Rev. 1992;16:219-233.[Medline] [Order article via Infotrieve]
21. Barone FC, Schmidt DB, Hillegrass LM, Price WJ, White RF, Feuerstein GZ, Clark RK, Lee EV, Griswold DE, Sarau HM. Reperfusion increases neutrophils and LTB₄ receptor binding in rat focal ischemia. Stroke. 1992;23:1337-1348.[Abstract/Free Full Text]
22. Bederson JB, Pitts LH, Nishimura MC, Davis RL, Bartkowksi H. Rat middle cerebral artery occlusion: evaluation of the model and development of a neurological examination. Stroke. 1986;17:472-476.[Abstract/Free Full Text]
23. Bederson JB, Pitts LH, German SM, Nishimura MC, Davis RL, Bartkowski HM. Evaluation of 2,3,5-tetraphenyltetrazolium chloride as a stain for detection and quantification of experimental cerebral infarction in rats. Stroke. 1986;17:1304-1308.[Abstract/Free Full Text]
24. Jiang JY, Lyeth BG, Clifton GL, Jenkins LW, Hamm RJ, Hayes RL. Relationship between body and brain temperature in traumatically brain-injured rodents. J Neurosurg. 1991;74:492-496.[Medline] [Order article via Infotrieve]
25. Barone FC, Feuerstein GZ, White RF. Brain cooling during transient focal ischemia provides complete neuroprotection. Neurosci Biobehav Rev. 1997;21:31-44.[Medline] [Order article via Infotrieve]
26. Novelli A, Reilly JA, Lysko PG, Henneberry RC. Glutamate becomes neurotoxic via the N-methyl-D-aspartate receptor when intracellular energy levels are reduced. Brain Res. 1988;451:205-212.[Medline] [Order article via Infotrieve]
27. Lysko PG, Cox JA, Vigano MA, Henneberry RC. Excitatory amino acid neurotoxicity at the N-methyl-D-aspartate receptor in cultured neurons: pharmacological characterization. Brain Res. 1989;499:258-266.[Medline] [Order article via Infotrieve]
28. Manev H, Bertolino M, DeErausquin G. Amiloride blocks glutamate-operated cationic channels and protects neurons in culture from glutamate-induced death. Neuropharmacology.. 1990;29:1103-1110.[Medline] [Order article via Infotrieve]
29. Lysko PG, Webb CL, Feuerstein G. Neuroprotective effects of carvedilol, a new antihypertensive, as a Na⁺ channel modulator and glutamate transport inhibitor. Neurosci Lett. 1994;171:77-80.[Medline] [Order article via Infotrieve]
30. Lysko PG, Webb CL, Yue T-L, Gu J-L, Feuerstein G. Neuroprotective effects of tetrodotoxin as a Na⁺ channel modulator and glutamate release inhibitor in cultured rat cerebellar neurons and in gerbil global brain ischemia. Stroke. 1994;25:2476-2482.[Abstract]
31. Lysko PG, Lysko KA, Yue T-L, Webb CL, Gu J-L, Feuerstein G. Neuroprotective effects of carvedilol, a new antihypertensive, in cultured rat cerebellar neurons and in gerbil global brain ischemia. Stroke. 1992;23:1630-1636.[Abstract/Free Full Text]
32. Yue TL, McKenna PJ, Lysko PG, Gu JL, Lysko KA, Ruffolo RR Jr, Feuerstein G. SB 211475, a metabolite of carvedilol, a novel antihypertensive agent, is a potent antioxidant. Eur J Pharmacol. 1994;251:237-243.[Medline] [Order article via Infotrieve]
33. Liu T, McDonnell PC, Young PR, White RF, Siren AL, Barone FC, Feuerstein GZ. Interleukin-1ß expression in ischemic rat cortex. Stroke. 1993;24:1746-1751.[Abstract/Free Full Text]
34. Latini R, Bianchi M, Correale C, Dinarello CA, Fantuzzi G, Fresco C, Maggioni AP, Mengozzi M, Romano S, Shapiro L, Sironi M, Tognoni G, Turato R, Ghezzi P. Cytokines in acute myocardial infarction: selective increases in circulating tumor necrosis factor, its soluble receptor, and interleukin-1 receptor antagonist. J Cardiovasc Pharmacol. 1994;23:1-6.[Medline] [Order article via Infotrieve]
35. Mohler KM, Torrance DS, Smith CA, Goodwin RG, Stremler KE, Fung VP, Madani H, Widmer MB. Soluble tumor necrosis factor (TNF) receptors are effective therapeutic agents in lethal endotoxemia and function simultaneously as both TNF carriers and TNF antagonists. J Immunol. 1993;151:1548-1561.[Abstract]
36. Nakane A, Numata A, Minagawa T. Endogenous tumor necrosis factor, interleukin-6 and gamma interferon levels during Listeria monocytogenes infection in mice. Infect Immun.. 1992;60:523-528.[Abstract/Free Full Text]
37. Sheehan KC, Ruddle NH, Schreiber RD. Generation and characterization of hamster monoclonal antibodies that neutralize murine tumor necrosis factors. J Immunol. 1989;142:3884-3894.[Abstract]
38. Ruddle NH, Bergman CM, McGrath KM, Lingenheld EG, Grunnet ML, Padula SJ, Clark RB. An antibody to lymphotoxin and tumor necrosis factor prevents transfer of experimental allergic encephalomyelitis. J Exp Med. 1990;172:1193-1200.[Abstract/Free Full Text]
39. Engelmann H, Novick D, Wallach D. Two tumor necrosis binding proteins purified from human urine: evidence for immunological cross-reactivity with cell surface tumor necrosis factor receptors. J Biol Chem. 1990;265:1531-1542.[Abstract/Free Full Text]
40. Schall TJ, Lewis M, Koller KJ, Lee A, Rice GC, Wong GH, Gatanaga T, Granger GA, Lentz R, Raab H. Molecular cloning and expression of a receptor for human tumor necrosis factor. Cell. 1990;61:361-370.[Medline] [Order article via Infotrieve]
41. Shohami E, Bass E, Wallach D, Yamin A, Gallily R. Inhibition of tumor necrosis factor alpha (TNF) activity in rat brain is associated with cerebroprotection after closed head injury. J Cereb Blood Flow Metab. 1996;16:378-384.[Medline]